Here is a thought experiment: Follow two groups until the
Survivorship is ZERO in both groups. Clearly, there is no
difference in "final" outcome.
But there still may be a HUGE difference in how fast they
reach zero.
Presumably, this sort of contrast is the source of the
difference in your two tests outcomes.
--
Rich Ulrich
Hello
I am running a survival analysis. Initially I compared the proportion of patients that reached the event , between the groups (using chi square) and there was no significance. But when I plotted the kaplan meier curves and run the log-rank there was a significance.
Does this mean anything? Should I defer the chi square as inappropriate for survival analysis and focus only on the log rank?
Many thanks
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
